%0 Conference Proceedings
%4 sid.inpe.br/sibgrapi/2013/01.24.14.52
%2 sid.inpe.br/sibgrapi/2013/01.24.14.52.38
%@isbn 978-85-7669-272-0
%T Microcanonical optimization applied to visual processing: an assessment
%D 1994
%A Torreão, José Ricardo A.,
%A Roe, Edward,
%@affiliation Departamento de Informática da Universidade Federal de Pernambuco (UFPE)
%@affiliation Departamento de Informática da Universidade Federal de Pernambuco (UFPE)
%E Freitas, Carla dal Sasso,
%E Geus, Klaus de,
%E Scheer, Sérgio,
%B Simpósio Brasileiro de Computação Gráfica e Processamento de Imagens, 7 (SIBGRAPI)
%C Curitiba, PR, Brazil
%8 9-11 Nov. 1994
%I Sociedade Brasileira de Computação
%J Porto Alegre
%V 1
%P 221-228
%S Anais
%K visual processing, microcanonical optimization algorithm, Barnard’s microcanonical.
%X The microcanonical optimization algorithm (µO) is an algorithm based on the microcanonical simulation techniques of statistical physics, which has been recently proposed as a suitable alternative to the simulated annealing approaches. The µO algorithm retains all of the positive features of the microcanonical annealing introduced by Barnard (which follows the pioneering work by Creutz), but avoids the necessity of an annealing schedule, thus resulting in improved computational efficiency. Here we present applications of the µO algorithm to some prototypical visual and show that it performs better than the traditional, or canonical, simulated annealing (SA) and then Barnards microcanonical annealing (MA).
%9 Métodos e Técnicas em Processamento Visual
%@language en
%3 29 Microcanonial optimization applied to visual.pdf